The scammer knew you better than your own family. That is why you would have fallen for it.
How leaked data, OSINT, and modern social engineering allow criminals to know everything about you before the first contact. The scam is not improvised. It is researched.
We grow up imagining scams as improvised. Some stranger calls, makes up a story, throws it out there to see if it sticks. People who do not fall for it are smart. People who do are distracted.
In 2026, that is folklore. The modern scam does not start with the phone call. It starts three weeks earlier, with someone studying you. Reading your Instagram. Crossing your leaked CPF with public databases. Finding your mother's name on her Facebook page. Identifying your main bank by the color of the card that appeared in one of your photos. Knowing where your daughter goes to school because you posted a graduation photo. Knowing you were abroad in January because you checked in.
When the phone call happens, the scammer knows you in a way that even your closest family probably does not. And that is what makes victims fall: not the victim's naivety, but the attacker's sophistication.
Let us reconstruct a typical 2026 scam
Any person. Let us call her Helena, 43 years old, a teacher at a private school in Belo Horizonte. Married to Roberto, a sales manager. Two children, a teenager and an 8-year-old.
Helena has never fallen for a scam. She knows banks do not ask for passwords. She does not click on suspicious SMS links. When she gets a call from an unknown number, she usually hangs up quickly.
Here is how a professional scammer would build their approach to her in 2026.
Week 1: collection
Helena's CPF leaked in at least three large public databases (Serasa 2021, Caixa 2023, an e-commerce platform in 2024). On criminal digital forums, these CPFs are sold in packages for R$ 10 to R$ 50 each, along with name, address, date of birth, phone number, and email.
The scammer buys Helena's package for R$ 30 and receives:
- Full name, CPF, RG, date of birth
- Full address, landline, mobile number
- Primary email
- Estimated income (from scoring databases)
- Approximate credit score
- Banks where she has a relationship (cross-referencing card databases)
Week 2: OSINT enrichment
With the full name, the scammer searches social media. In 20 minutes:
- Helena's Instagram (public): 847 photos. Shows the children, the husband, the dog, the school where she works, recent trips (Maceió in January, Buenos Aires last July).
- Helena's Facebook (public to friends, but the scammer creates a fake account): birthday, mother ("My darling daughter whom I carry in my heart"), workplace, education.
- LinkedIn: previous jobs, education, precise dates, professional network.
- Twitter/X (semi-public): political opinions, hobbies, frustrations with banks.
Sees the husband, discovers he has a WhatsApp Business account (sales manager), finds the company he works for. Sees a recent photo of Helena with a card visible on the table: Banco do Brasil.
Cross-referencing everything: Helena 43 years old, married to Roberto 47 (sales manager at [company]), two children (Lucas teenager, Manuela 8 years old), school where she works, Banco do Brasil customer, estimated income R$ 8,000 to R$ 12,000, travels frequently, has a credit card, dog named Tofu.
Total time invested: 90 minutes. Cost: R$ 30 plus research time.
Week 3: opportunity trigger
The scammer monitors Helena's social media for a few days, waiting for a moment of vulnerability. Friday morning, Helena posts a photo: "Heading to the parent meeting tomorrow, anxiety through the roof, can anyone explain how to raise a teenager?" Family engagement, comments.
The scammer now knows:
- Saturday morning, Helena will be emotionally engaged (son, school)
- She is anxious, distracted
- She will probably have her son nearby, not alone at home with time to think
Saturday morning, 10:14 a.m. Helena receives a call from a regular mobile number (not 0800). She answers.
The contact
"Helena? This is Mariana from Banco do Brasil's customer center. We detected a suspicious transaction on your account just now, in the amount of R$ 4,876.32. Did you authorize a purchase at Magazine Luiza at 9:47 this morning?"
Helena, surprised: "No, I did not."
"I understand, ma'am. This looks like fraud. I am going to block the transaction now, but I need to verify a few security details with you quickly before the blocking window expires. May I?"
Helena: "Of course."
"For security, I need you to confirm: Helena Aparecida Soares Pereira, CPF 423.xxx.xxx-xx, date of birth March 14, 1981, address Rua das Acácias 234 apartment 502, Pampulha neighborhood. Is all of that correct?"
Helena, pleasantly surprised because all the details are accurate (it must really be the bank): "Yes, that is all correct."
"Perfect. To block the transaction, I need you to confirm: was the R$ 4,876 transaction made on the card ending in xxxx or xxxx? I can see both of your cards here."
Helena checks the numbers (her own actual cards, from leaked databases) and answers correctly.
"OK, that is the one. To complete the block, I am sending you an SMS code right now to your mobile. When it arrives, please read me the 6 digits to confirm authentication on our end. It is not for anyone to open, just for me to confirm I am actually speaking with Helena."
Helena receives an SMS: "Banco do Brasil. Your authentication code is 487391. NEVER share this code with anyone."
Helena reads the SMS. She hesitates for 2 seconds. The bank's "Mariana" says: "It is just to confirm I am speaking with you. Can you read it?"
Helena reads: "487391."
"Mariana" continues: "Confirmed. Block completed. You will receive a receipt by email within the next two hours. Is there anything else?"
Helena thanks her. She hangs up, relieved.
Thirty minutes later, R$ 14,300 had left Helena's account via Pix to a mule account. The SMS code she read was not a block confirmation. It was the Pix authorization code for the transfer.
Why Helena fell (and she is not "stupid")
Every element of the scam was designed to reduce Helena's chances of suspecting anything:
1. A regular mobile number, not 0800. Modern banks use a mix. Helena has no clear rule about this.
2. Mariana knew the full name, CPF, address, and card numbers. All true. In a fraction of a second, Helena's brain interpreted this as "this person has access to my internal data, so she must be from the bank."
3. Manufactured urgency. "Suspicious transaction right now" plus "blocking window expiring" create time pressure. Under pressure, critical thinking is abandoned.
4. Helena was mentally somewhere else. Saturday morning, anxious about the parent meeting, children nearby. Reduced capacity for analysis.
5. The story made sense. Fraud is common. Banks proactively blocking transactions is common. A proactive call is rare but plausible. Thinking "this is a scam" requires a pause Helena did not take.
6. Mariana asked her to read a code that seemed to be for authentication. Helena saw the warning "NEVER share this code" but "Mariana" deflected its meaning ("it is just to confirm I am speaking with you"). Helena followed the normal social script of being helpful.
Helena is not careless. She had the bad luck of encountering a professional scam, built on three weeks of specific research about her, at a moment when she was distracted.
The ingredient that changes everything: leaked data
The foundation of the scam above is not the phone call. It is the R$ 30 package the scammer bought. Without that data, the phone call would not have convinced Helena for even 10 seconds. With it, the scam gained instant credibility.
In 2026, more than 75% of Brazilians have had personal data leaked in at least one public criminal database. The leaks came from:
- Banks and fintechs (multiple institutions over the past 5 years)
- E-commerce platforms (Magazine, Americanas, others)
- Telecom companies (Claro, Vivo, TIM in separate incidents)
- Popular apps (delivery, ride-sharing)
- Data brokers (Serasa, SPC)
The data keeps circulating on forums for years. It gets updated with new leaks. It gets more detailed. In some packages, you can buy a specific person's transaction history.
The scammer who calls is not guessing. They are reading.
How to protect yourself (without becoming paranoid)
You cannot "un-leak" data that is already out there. But you can make social engineering harder to build on top of it.
1. Reduce social media exposure
- Lock Instagram and Facebook to real friends. A public account is free ammunition for the scammer.
- Do not post photos of your card, even blurred (forensic analysis can recover it).
- Do not post photos of your home, neighborhood, or children's school.
- Do not post while you are traveling. Post when you get back.
2. Change the "verification" dynamic with your bank
Good Brazilian banks allow you to configure an alternative contact method. Instead of the bank calling you (a vector for scams), you call the bank when something seems suspicious. Teach your family this inversion: "if someone calls saying they are from the bank, thank them, hang up, and call the official number on the back of your card."
3. Passwords and codes are NEVER authenticated by voice
Absolute rule: no bank, under any circumstances, will ask you to read an SMS code over the phone. SMS codes are authorizations for something, not identity verification.
Same rule: no bank asks for your app password, card PIN, internet banking password, 6-digit code, transfer code, or Pix code. Nothing. Full stop.
Repeat this to your mother, your grandmother, your teenage children who are starting to have digital bank accounts. Print it and put it on the wall if needed.
4. Reduce data overlap in low-quality apps
The fewer registrations you have with CPF, address, phone number, and card details, the less data there is to leak. Use unique passwords for every service (a personal digital vault solves this). Use a throwaway email for non-critical registrations. Consider a secondary phone number (a R$ 30/month SIM) for suspicious signups.
5. Encrypted client-side personal digital vault
This is where TAIVA Vault comes in directly. When you use an encrypted client-side vault:
- Your passwords are unique per service (automatic generator). If one leaks, the other 200 are untouched.
- The vault stores your sensitive data (CPF, ID, cards) encrypted. Even the TAIVA server cannot read it. A TAIVA data breach does not expose you.
- 2FA TOTP codes are generated in the app without depending on SMS (which can be intercepted via SIM swap).
- Scanned documents (ID, driver's license) are encrypted and available when you legitimately need them, not floating around in email.
The vault does not prevent the scammer from buying your old data. But it dramatically reduces how much new data leaks going forward, and it keeps passwords unique per service (cutting the spread of a single leak).
6. Set up alerts
Good Brazilian banks allow push and SMS notifications for every transaction. Enable them on all accounts. For any charge, you see it immediately. If something strange appears, call the bank at the number on the back of your card, not the number that called you.
7. Educate family
Parents, grandparents, teenage children. Social engineering works at any age. A 20-minute conversation with your family about how scams work is worth more than any security app. Tell real cases. Show examples. Agree on safe code words to confirm identity in suspicious situations.
The simple rule
The modern scam is not stupidity on the part of the victim. It is research on the part of the attacker, with abundant leaked data tools, and a carefully designed emotional architecture to bypass critical thinking in perfectly intelligent people.
The defense is not "be smarter." It is reducing the attack surface: less data exposed on social media, unique passwords in a personal digital vault, 2FA on authenticator apps (not SMS), inverted contact channel with the bank, and a family educated on the rules.
And accepting the uncomfortable truth: the scammer who will call you in 6 months is probably already collecting data about you right now. The question is not whether the attack will happen. It is whether you will be prepared when it comes.
TAIVA Vault: personal digital vault with automatically generated unique passwords, integrated TOTP 2FA, encrypted storage for ID/driver's license/cards, post-quantum cryptography. It does not prevent past leaks, but it dramatically reduces your future attack surface. Create free account →
This article is informational. The case described is a composite based on real scam patterns documented in Brazil; names and specific details are fictional. If you have been the victim of a scam, contact your bank immediately and file a police report (via 147 or the online police station).
Enjoyed the article?
TAIVA is a post-quantum password vault hosted in Brazil. Your passwords are encrypted client-side, with your key split between two servers. Free forever, PRO at R$49/mo.